Improved Deep Embeddings for Inferencing with Multi-Layered Networks
This work addresses the challenge of exploiting shared and distinct information in multi-layered networks for researchers and practitioners in network analysis, representing an incremental improvement over existing embedding methods.
The paper tackles the problem of inferencing with multi-layered networks by proposing a novel approach that jointly obtains node embeddings via DeepWalk on a supra graph and fine-tunes them to encourage cohesive structure, resulting in outperformance over existing algorithms in node classification, link prediction, and multi-layered community detection on several benchmarks, with effective scaling up to 37 layers and consistent production of highly modular community structure.
Inferencing with network data necessitates the mapping of its nodes into a vector space, where the relationships are preserved. However, with multi-layered networks, where multiple types of relationships exist for the same set of nodes, it is crucial to exploit the information shared between layers, in addition to the distinct aspects of each layer. In this paper, we propose a novel approach that first obtains node embeddings in all layers jointly via DeepWalk on a \textit{supra} graph, which allows interactions between layers, and then fine-tunes the embeddings to encourage cohesive structure in the latent space. With empirical studies in node classification, link prediction and multi-layered community detection, we show that the proposed approach outperforms existing single- and multi-layered network embedding algorithms on several benchmarks. In addition to effectively scaling to a large number of layers (tested up to $37$), our approach consistently produces highly modular community structure, even when compared to methods that directly optimize for the modularity function.